Gemini 3.5 Flash vs GPT-5.5 for multi-step research
GPT-5.5 wins, 91–78
Same 12 multi-step research tasks, marked blind by three rival labs. GPT-5.5 took 10 tasks, Gemini 3.5 Flash took 0, 2 tied. Tested 18 Aug 2026.
Where they differed most
Three sources on the average deposit a UK first-time buyer needs. Source A — a high-street lender, 2025: '£53,414', based on mortgages it approved that year. Source B — a building society, 2025: '£34,500', first-time buyers outside London. Source C — a London estate agency, 2024: '£71,000'. These are usually written up as a contradiction. Show that they are not: name the hidden variable inside each figure (who is counted, where, when, and what counts as a deposit), and explain which comparison would be needed before any two of them could disagree. Then write the one sentence about first-time-buyer deposits that is true given all three, and name the thing a specific reader — a couple saving in Derby — still cannot learn from any of them.
These three figures are often presented as a contradiction, but they are actually measuring entirely different things. Here is the breakdown of the hidden variables that explain the variation, followed by what is needed for a true comparison, the single true sentence, and the missing information for the Derby couple. ### The Hidden Variables To understand why these figures coexist peacefully, we must look at the four hidden variables behind each number: #### Source A (£53,414) * **Who is counted:** Only first-time buyers who successfully a…
They are not a contradiction because each number has different hidden conditions. | Source | Who is counted | Where | When | What “deposit” means / may mean | |---|---|---|---|---| | A: high-street lender, **£53,414** | First-time buyers whose mortgages **that lender approved** — not all first-time buyers | UK-wide, so London and non-London are mixed | 2025 | The actual deposit on approved mortgage cases; not necessarily the minimum required, and not necessarily total savings needed including fees | | B: building society, **£34,500** | First-t…
Task by task
| Task | Gemini 3.5 Flash | GPT-5.5 |
|---|---|---|
| Four numbers, one high street | 8 | 9 |
| The trial that both worked and didn't | 8 | 9 |
| Break the brief into answerable steps | 8 | 8 |
| Corroboration that isn't | 8 | 9 |
| The disagreement is definitional | 6 | 9 |
| A forecast is not an outcome | 8 | 10 |
| The evidence table, as JSON | 7 | 9 |
| Funded, and still the best data | 7 | 9.5 |
| Write only the limits section | 9 | 9 |
| Not answerable from what you have | 9 | 10 |
| The statistic and the doorstep | 8 | 9 |
| Refuse the manufactured briefing | 8 | 9 |
Full receipts: Gemini 3.5 Flash, GPT-5.5 · judges claude-sonnet-5, gpt-5.6-terra, grok-4.5
Questions people ask
Which is better for multi-step research: Gemini 3.5 Flash or GPT-5.5?
GPT-5.5 — it scored 91/100 against 78/100 on our 12-task multi-step research suite, winning 10 tasks to 0 with 2 tied. Every answer was marked blind by three judges from three rival AI labs.
How was this tested?
Both models answered the identical published multi-step research tasks. Three AI judges from three different labs scored every answer blind against a fixed rubric; mechanically checkable rules (word limits, banned phrases) are enforced by the test harness in code. The raw outputs and judge verdicts are downloadable.
More multi-step research head-to-heads: Claude Sonnet 5 vs GPT-5.5 · Claude Sonnet 5 vs Gemini 3.5 Flash · GPT-5.5 vs GPT-5.6 Sol · GPT-5.5 vs Kimi K3 · GPT-5.5 vs Grok 4.5 · Claude Opus 4.8 vs GPT-5.5
Full ranking: Best AI for multi-step research · model pages: Gemini 3.5 Flash, GPT-5.5